|
--- |
|
library_name: keras-hub |
|
license: apache-2.0 |
|
tags: |
|
- image-classification |
|
pipeline_tag: image-classification |
|
--- |
|
## Model Overview |
|
The VGG model is a type of convolutional neural network (CNN) architecture designed for image recognition and classification tasks. Developed by the Visual Geometry Group at the University of Oxford, it was introduced in the paper titled "Very Deep Convolutional Networks for Large-Scale Image Recognition" by Karen Simonyan and Andrew Zisserman in 2014. This model is supported in both KerasCV and KerasHub. KerasCV will no longer be actively developed, so please try to use KerasHub. |
|
|
|
|
|
|
|
## Links |
|
* [VGG paper](https://arxiv.org/abs/1409.1556) |
|
|
|
## Installation |
|
|
|
Keras and KerasHub can be installed with: |
|
|
|
``` |
|
pip install -U -q keras-Hub |
|
pip install -U -q keras>=3 |
|
``` |
|
|
|
Jax, TensorFlow, and Torch come preinstalled in Kaggle Notebooks. For instructions on installing them in another environment see the [Keras Getting Started](https://keras.io/getting_started/) page. |
|
|
|
## Presets |
|
|
|
The following model checkpoints are provided by the Keras team. Weights have been ported from https://huggingface.co/timm. |
|
|
|
| Preset Name | Parameters | Description | |
|
|------------------|------------|----------------------------------------------------------------| |
|
| vgg_11_imagenet | 9.22M | 11-layer VGG model pre-trained on the ImageNet 1k dataset at a 224x224 resolution. | |
|
| vgg_13_imagenet | 9.40M | 13-layer VGG model pre-trained on the ImageNet 1k dataset at a 224x224 resolution. | |
|
| vgg_16_imagenet | 14.71M | 16-layer VGG model pre-trained on the ImageNet 1k dataset at a 224x224 resolution. | |
|
| vgg_19_imagenet | 20.02M | 19-layer VGG model pre-trained on the ImageNet 1k dataset at a 224x224 resolution. | |
|
|
|
## Example Usage |
|
```python |
|
input_data = np.ones(shape=(2, 224, 224, 3)) |
|
|
|
# Pretrained backbone |
|
model = keras_hub.models.VGGBackbone.from_preset("vgg_16_imagenet") |
|
model(input_data) |
|
|
|
# Randomly initialized backbone with a custom config |
|
model = keras_hub.models.VGGBackbone( |
|
stackwise_num_repeats=[2, 3, 3, 2], |
|
stackwise_num_filters=[64, 128, 256, 512], |
|
) |
|
model(input_data) |
|
|
|
# Use VGG for image classification task |
|
model = keras_hub.models.ImageClassifier.from_preset("vgg_16_imagenet") |
|
|
|
# User Timm presets directly from HuggingFace |
|
model = keras_hub.models.ImageClassifier.from_preset('hf://timm/vgg11.tv_in1k') |
|
``` |
|
|
|
## Example Usage with Hugging Face URI |
|
|
|
```python |
|
input_data = np.ones(shape=(2, 224, 224, 3)) |
|
|
|
# Pretrained backbone |
|
model = keras_hub.models.VGGBackbone.from_preset("hf://keras/vgg_16_imagenet") |
|
model(input_data) |
|
|
|
# Randomly initialized backbone with a custom config |
|
model = keras_hub.models.VGGBackbone( |
|
stackwise_num_repeats=[2, 3, 3, 2], |
|
stackwise_num_filters=[64, 128, 256, 512], |
|
) |
|
model(input_data) |
|
|
|
# Use VGG for image classification task |
|
model = keras_hub.models.ImageClassifier.from_preset("hf://keras/vgg_16_imagenet") |
|
|
|
# User Timm presets directly from HuggingFace |
|
model = keras_hub.models.ImageClassifier.from_preset('hf://timm/vgg11.tv_in1k') |
|
``` |
